PROJECT A Although the locations and relative activities of recombination hotspots are crucial in determining patterns of inheritance from one generation to the next, as yet we have only a rudimentary understanding of how these parameters are determined. Allelic variation in PRDM9 now provides us the opportunity to remedy this deficiency. The long-term goal of this project is to characterize the molecular and genetic mechanisms that regulate hotspot location and activity. To accomplish this, we will generate a series of KO and knock-in mice to assure that the only genetic variable will be the identity of PRDM9 allele being tested. Using this constant genetic background, we will identify allele-specific hotspots and examine the role of allelic preference for hotspot activation when two alleles are present in the same meiotic cells. Using genetically variable backgrounds, we will also identify genetic modifiers of PRDM9 function. In Aim 1 we will: (a) identify a series of PRDM9 allele-specific hotspots whose sequences will help clarify the rules governing PRDM9 allelic specificity; (b) determine if there are autosomal as well as pseudo- autosomal PRDM9-independent hotspots, and (c) test using mice as a vehicle for characterizing human PRDM9 specificity. It is not possible to characterize the biological functions of human PRDM9 in depth using human material, as hotspots controlled by the same PRDM9 allele vary over fifteen fold in activity from one individual to the next, presumably from genetic variation at multiple other hotspot regulatory factors (4). Testing the human allele also provides a useful control by comparing recombination in the presence of a drastically different ZNF domain. In doing so, as outlined in the Introduction, we believe that it is important to assay both the DSBs that characterize the beginnings of the recombination process and the genetic crossovers that are the final, functional outcome. The two measurements provide complementary insights. If we want to understand the functional consequences of recombination, we do best to provide descriptions of both the beginning and end of the process. In Aim 2 we expect to confirm the existence of allelic preference in the ability of PRDM9 alleles to activate recombination, and then test whether susceptibility to this effect depends on the relative strength of the hotspots themselves, and importantly determine if preference is mediated by the relative numbers of PRDM9 molecules present or their relative affinity for hotspot DNA sequences. These experiments have considerable relevance for understanding the phenomenon of crossover homeostasis that operates within constraints on the numbers of DSBs to assure a constant number of crossovers at each meiosis. Finally, in Aim 3 we will test a variety of mouse genetic backgrounds to screen for the existence of genetic modifiers of PRDM9 function with the goal of determining their molecular identity. PRDM9 cannot act in isolation, and the goal here is to identify proteins that physically or functionally interact with PRDM9 The knock-in strains of PRDM9 alleles generated in this Project will be used in Project B Petkov and Project C Hibbs. Carrying out these goals has significance both for understanding the basic biology of meiosis and for its public health implications, as failures of meiotic recombination and gametogenesis are major contributors to human in fertility and embryonic lethality.